Ground surface multi-modal inspection robot

Abstract

A system for inspecting surfaces that includes a mobile base, sensors for base navigation, sensors for surface inspection, a communication system and a host computer that executes modules for base motion planning and navigation, location, point cloud acquisition and processing, surface modelling and analysis, multi module coordination and user interfaces. The inspection procedure has the robot move in a zigzag pattern trajectory over the surface. For every fixed distance, a 3D point cloud of the surface is generated and the location of the point cloud with respect to the world coordinate system is recorded. The location of the point cloud is based on SLAM for spatial mapping. At the same time, a high-resolution photo of the corresponding area on the surface is recorded by the camera. Both the point cloud and the photo are transmitted to the host computer for processing and analysis. This information is used in a new 3D detection and image processing algorithm to find flaws in the surface like bumps or depressions. If irregular flaws are detected, the robot marks such a problematic location.

1 . A system for inspecting surfaces comprising:
 a mobile base that moves over the surface,   sensors for base navigation located on the base,   sensors for surface inspection located on the base that detect defects in the surface in the form of bumps in the surface and depressions in the surface greater than a threshold value,   a communication system located on the base′   a robot computer located on the base for controlling the sensors and communication system, and   a host computer that executes modules for base motion planning and navigation, position location, point cloud acquisition and processing, surface modelling and analysis, multi module coordination and user interfaces,   wherein the host computer delivers commands to the robot computer and receives signals from the robot computer through the communication system.   
     
     
         2 . The system of  claim 1  wherein the inspection procedure has the following steps:
 the robot moves in a zigzag pattern trajectory over the surface; 
 for every fixed distance, a 3D point cloud of the surface is generated and the location of the point cloud with respect to the world coordinate system is recorded, the location of the point cloud is based on SLAM for spatial mapping; 
 at the same time, a high-resolution photo of the corresponding area on the surface is recorded by the camera; 
 both the point cloud and the photo are transmitted to the host computer for processing and analysis, which information is used in a new 3D detection and image processing algorithm to find flaws in the surface like bumps or depressions; and 
 if irregular flaws are detected, the robot marks such a problematic location. 
 
     
     
         3 . The system of  claim 2  wherein the robot moving in the zigzag pattern trajectory comprises the robot following steps of:
 moving forward along a first scan row to the end, 
 reversing its direction without turning around and moving backward at an angle to the second scan row, 
 moving backwards along the second scan row to the end, 
 reversing its direction without turning around and moving forward at an angle to the third row, and 
 moving forward along the third scan row to the end. 
 
     
     
         4 . The system of  claim 1  wherein the sensors for navigation are based on lidar positioning. 
     
     
         5 . The system of  claim 1  wherein the sensors for navigation are based on Global Navigation Satellite System (GNSS). 
     
     
         6 . The system of  claim 5  wherein the GNSS comprises:
 a reference station of the GNSS with known coordinates that receives signals from a satellite of the GNSS; 
 an antenna mounted on the mobile base that acts as a rover station that directly receives signals from the same satellite as the reference station, whereby through differential operation the antenna and reference station eliminate shared errors and obtain a more accurate positioning solution; and 
 wherein the corrected differential signals are transmitted to the GNSS position location module. 
 
     
     
         7 . The system of  claim 1  further including camera for taking images of defects located in the surface. 
     
     
         8 . The system of  claim 1  further including spray device for spraying an erasable paint or chalk on the surface to mark defects located in the surface. 
     
     
         9 . The system of  claim 1  wherein the sensors for surface inspection comprise:
 a line projecting laser that projects a laser line from a bracket on and above the base vertically downward toward the surface; and 
 a 3D camera located on the base and imaging the laser line at an angle, said camera analyzing an image of the laser line to locate the laser line and determine any change in the surface based on its image of the laser line. 
 
     
     
         10 . The system of  claim 1  further including an inertial measurement unit (IMU) in the form of a gyroscope mounted on the base that obtains robot position and attitude data during its movement over the surface. 
     
     
         11 . The system of  claim 1  further including at least one ultrasonic safety protection sensor to detect obstacles and prevent collisions with the base. 
     
     
         12 . The system of  claim 11  wherein ultrasonic safety protection sensors are installed on all sides of the mobile base.